INTRODUCTION
The current water shortage in Spain is mainly due to the poor management of this resource in agriculture. Though it represents only 3% of the GDP, it accounts for nearly 80% of Spanish water consumption. Therefore, problems inherent to water shortage in Spain could be solved to a great extent by applying efficient agricultural water use practices. In this context, our purpose is to relate environmental impact and water use efficiency to the system of relationships and organization in the agricultural sector. Specifically, we considered family farms in the province of Almería in southeast Spain, as our reference case. Their use of water is highly efficient compared to other Spanish agro-food systems. Recent analyses show that the environmental impact of water use in horticulture in Almería, as measured by its water footprint (WF), is 20 times lower than the mean in the rest of Spanish agriculture (Sotelo, Reference Sotelo2011), and water use is six times lower than farms in the rest of Spain (Fernández et al., Reference Fernández, González, Carreño, Pérez and Bonachela2007). For this reason, it is important to examine facets of these family structures that have a bearing on better water use, in order to apply the results to other regions and/or countries.
The agricultural system in the province of Almería in southeast Spain, known as the ‘Almería model’ (Aznar-Sánchez and Picón, Reference Aznar-Sánchez and Picón2010; Ferraro-García et al., Reference Ferraro-García, Aznar-Sánchez, Mesa-Barreto and Aguilera Díaz2000; Galdeano-Gómez et al., Reference Galdeano-Gómez, Aznar-Sánchez and Pérez-Mesa2011; Molina-Herrera, Reference Molina-Herrera2005), has undergone unprecedented transformation in the recent history of this country. Its success is due to the intensive farming of fruits and vegetables in a productive organization based on small family farms, the local marketing structure composed mainly by cooperatives, and related secondary industries. Environmental components have also played a major role, particularly in the efficient use of natural resources, as the production system has become more and more respectful of its surroundings (Galdeano-Gómez et al., Reference Galdeano-Gómez, Céspedes-Lorente and Martínez-Del Río2008) due to the technology employed in horticulture.
The first feature of this farming system that should be considered is its origins relied on the exploitation of underground aquifers that were sufficient to supply its early development in this semi-arid province, which is practically a desert. The initial technological step was to introduce the technique known as ‘sandy soil’, which consists of covering the soil with a layer of sand, thus transforming unproductive land into prosperous farms with larger yields in winter. In the 1960s, the next major technological improvement introduced was the greenhouse, built to protect the crops from wind and low winter temperatures, but which also contributed to water conservation. In the 40 years since then, the area devoted to greenhouse cultivation in the province has undergone spectacular growth, from 3000 ha in 1970 to 29 035 in 2014 (Fundación Cajamar, 2014), making it the province with the largest area of greenhouse crops in Spain. This structure has reduced erosion caused by the strong winds and occasionally heavy rainstorms characteristic of semi-arid regions.
At first, the enormous expansion of the area under cultivation generated strong pressure on available water, and some symptoms of aquifer degradation appeared, such as salinization near the coast. In the late 80s, some solutions to this problem began to be introduced, such as systems for saving water (drip irrigation, control of demand of water from aquifers, etc.) and increasing its availability (desalination plants, reservoirs and reuse). Sandy soil was also replaced by non-soil cultivation or hydroponic systems supplemented by efficient water use techniques. Since the early 2000s, the area devoted to production under greenhouses has not increased as rapidly because of increased costs and a trend towards stability of sale prices. This has led to a higher investment in technology to increase the productivity of crops and optimize resource efficiency. To improve water use, growers have installed several structures for collecting and storing rainwater. These changes are also promoted by farmers’ organizations, mainly cooperatives (marketing, financing, inputs supply) and irrigation communities, which play an important role in several agricultural changes (Galdeano-Gómez et al., Reference Galdeano Gómez, Pérez-Mesa and Godoy-Durán2016). As a result, Almería is currently the most efficient irrigation area in Spain, with wide use of drip irrigation systems and increased water treatment, recycling and reuse. Thus, water use in agriculture in Almería has a small impact on the environment, compared to the rest of Spain (Sotelo, 201; Tolón-Becerra et al., Reference Tolón-Becerra, Lastra Bravo and Fernández Membrive2013). This would not have come about without a change in attitude on the part of the growers, who are more ecologically aware than those in other places in Spain (Medina, Reference Medina2014).
Some previous studies have analysed the characteristics and behaviour of the family farms with regard to its capacity for innovation (Spriggs et al., Reference Spriggs, Yu, Deeds and Sorenson2012), environmental measures (Delmas and Gergaud, Reference Delmas and Gergaud2014; Medina, Reference Medina2014) or social responsibility (Cruz et al., Reference Cruz, Larraza-Kintana, Garcés-Galdeano and Berrone2014), but not its water use. Many studies (Duarte et al., Reference Duarte, Pinilla and Serrano2014; Hoekstra and Mekonnen, Reference Hoekstra and Mekonnen2012; Sotelo, Reference Sotelo2011; Tolón-Becerra et al., Reference Tolón-Becerra, Lastra Bravo and Fernández Membrive2013) have also calculated the WF for different countries, regions, sectors, etc., but few analyse the factors that influence water usage. This paper bridges a major gap in the literature related to lowering the environmental impact of water use and the family farm. To this aim, an empirical analysis was developed in order to evaluate the influence of socio-economic factors and management structure features of family farms on water use efficiency.
MATERIALS AND METHODS
Empirical setting and data analysis
A total of 55 intensive greenhouse family farms were chosen by random cluster sampling by area in the province of Almería, and they were surveyed during the 2014–2015 fruit and vegetable growing season (September to June). At the same time, interviews were carried out in irrigation communities in the province. The design of the surveys was composed by four dimensions, which enabled us to evaluate their approach to environmental awareness and efficient water use: current status of the family farm; fruit and vegetables grown by the family farm; environmental awareness and perspective on water use efficiency.
Irrigation communities were also interviewed to supplement the information collected in the grower surveys on: plans, improvements, innovations and new technologies for more efficient water use, and the type of support or government subsidies; and water use and management awareness programmes. When the surveys and interviews were completed, the information had to be validated to avoid any errors in their collection and analysis. Table 1 includes the information obtained from analysis of the dimensions that comprise the family farm.
Survey results.
a Growers who received subsidies or aids for improvement, innovation or new technology for reducing use of water.
Source: The authors, based on family business surveys.
Description of variables
Dependent variable
Efficiency indicates the efficiency of water usage, which was measured by the sum of three items (for a similar approach, see Tang et al., Reference Tang, Folmer and Xue2013): whether the family farm received environmental certification, and if so, to what extent this was related to efficient water use – scored from 0 (not certified or certification unrelated to water) to 5 (when the family farm was fully certified for water use efficiency); whether the family farm was carrying out any water use efficiency plan, dummy variable scored from 0 (No) to 1 (Yes); and whether the family farm had simply implanted some improvement, innovation or new technology for reducing water use, dummy variable scored 0 (No) and 1 (Yes). This was tested with the standardized variable, but the results did not vary substantially.
Independent variables
The determinants of whether the family farm is more or less aware of water vary considerably, but they usually include the characteristics of the farm, socio-demographic factors and environmental factors (see e.g., Jones and Dunlap, Reference Jones and Dunlap1992).
Control variables - family farm characteristics
- generation: Number of generations that have run the family farm, as an indication of its age. The generation, like other indicators of the family farm's age, is a variable commonly used as a control variable for analysing its behaviour (e.g., Delmas and Gergaud, Reference Delmas and Gergaud2014; Spriggs et al., Reference Spriggs, Yu, Deeds and Sorenson2012). Greater generational participation in a family farm may increase the complexity of its governing structures (Ling and Kellermanns, Reference Ling and Kellermanns2010), and thereby, decision-making on water efficiency.
- area: Number of square metres currently cultivated by the family farm as an indication of its size. The size of the family farm is also often used as a control variable (as in Delmas and Gergaud, Reference Delmas and Gergaud2014; Spriggs et al., Reference Spriggs, Yu, Deeds and Sorenson2012). According to McGrath (Reference McGrath2001), the size of the family farm could bias its capacity for innovation, for instance regarding water use efficiency.
-
– workers: Number of people working for the family farm, as another indication of its size.
-
– work_m2: Number of workers per square metre of area cultivated.
-
– sale_m2: Family farm sales per square metre of cultivated area.
-
– sales: Total family farm sales.
Variable related to who inherits the family farm
- inherit: Dummy variable scoring 1 if the farmer thinks the next generation will inherit the family farm or 0 when he does not. Delmas and Gergaud (Reference Delmas and Gergaud2014) also enter this variable as a determinant of environmental certification. The family farms make investments for long-term benefits to family members (Habbershon and Pistrui, Reference Habbershon and Pistrui2002). Taking steps for the efficient use of water enables growers to invest in long-term sustainability of their firm to the benefit of the next generation. On the contrary, the impossibility of passing on the business reduces the farmer's incentive for investing beyond his own life expectancy.
Variables related to family farm decision-makers
- dec.-makers: Number of people in the family farm making decisions. As Spriggs et al. (Reference Spriggs, Yu, Deeds and Sorenson2012), for example, conclude, the more there are, the more complicated making decisions on investing in improvements for more efficient water use becomes.
- age_under 45 years: Dummy variable, 1 if the mean age of decision-makers is under 45 or 0 when it is not. The figure of 45 was chosen as it is the farmers’ average age. In previous studies, results with respect to age have been ambiguous. Older growers have usually had more problems with water shortages than younger people, so from this point of view, the older the decision-makers are, the more aware of water the family farm would be expected to be (Lee and Zhang, Reference Lee and Zhang2008). However, younger growers could be more aware because they have a longer life expectancy and a longer time ahead of them to be earning income (Arcury and Christianson, Reference Arcury and Christianson1990). Therefore, the impact of age on awareness is an empirical question and no a priori hypothesis can be made on its impact.
- education: Average education of family farm decision-makers. The education of each was measured on a scale of 1 (no education), 2 (primary education), 3 (middle school), 4 (high school or vocational training) or 5 (university or higher education). Many studies have shown that more education usually makes individuals more aware of environmental problems in general (Jones and Dunlap, Reference Jones and Dunlap1992; Lee and Zhang, Reference Lee and Zhang2008). Therefore, we expected a positive relationship between the education of decision-makers and their water awareness.
- women: Number of woman decision-makers. According to Farmar-Bowers (Reference Farmar-Bowers2010), the contribution of female growers to strategic business decisions on sustainable development is very important. Therefore, family farms with more women decision-makers might be expected to be more aware of efficient water use.
Variables related to environmental and water awareness of the family farm and the surrounding area
- integrated_m2: Integrated Pest Management (IPM), organic and other certifications of environmental quality of production (in kilograms per square metre of total cultivated area). A weighted mean of all the crops was calculated.
- aid: Dummy variable of 1 if the family farm has received any government aid or subsidy for implanting their plans, improvements or innovations for using water more efficiently and 0 if none.
- environmental: This variable shows whether the main reason for the family farm using water more efficiently is environmental awareness and/or thinking of future generations. The variable is 0 if neither of these are reasons, 1 if one of them is a reason and 2 if both are reasons.
- economic: This variable shows whether the main reason for the family farm using water more efficiently is saving costs and/or increased sales. This variable is 0 if neither of these are its reasons, 1 if one of the two is and 2 if both.
- sector: This variable shows the farmer's evaluation of the importance of other family farms in the sector being aware of water use. Measured on a scale of 1 (unimportant) to 5 (very important).
- daily_life: How much the farmer tries to use less water in his daily life (e.g., taking a shower instead of a bath, recycling used water for other purposes, avoiding leaving taps open longer than necessary, etc.). Measured on a scale of 1 (not at all) to 5 (very much).
To make these variables easier to understand, Table 2 shows the descriptive statistics. Before entering the variables into the regression model, multicollinearity and normality have been studied (Table 3). There was only a high correlation between sales and area, but they are both control variables. The rest of the correlations were much lower, and so each variable can be said to represent a different concept.
Descriptive statistics of variables used.
Pearson correlations.
RESULTS
As in Spriggs et al. (Reference Spriggs, Yu, Deeds and Sorenson2012), a hierarchical regression model was used to test the influence of the different groups of independent variables on the efficiency of water usage. Thus, the variables were entered in four steps in the following sequence: model 1 – Control variables (generation, area, workers, work_m2, sales_m2 and sales); model 2 – The inherit variable was added to the above; Model 3 – The variables related to family farm decision-makers were added (decision-makers, age_under45, education and women); and Model 4 – The variables related to environmental and water awareness of the family farm (integrated_m2, aid, environmental, economic, sector and daily_life) were included. For each step, the variance explained (R 2), significance level (P value) and increase in R 2 and F values were evaluated (Table 4).
Results of hierarchical regression analysis.
Note: Standard errors in parentheses.
*P < 0.05. **P < 0.01. ***P < 0.001.
Model 1 includes the control variables and shows that the family farms that have been passed down through more generations, and are therefore older, take more measures to use water more efficiently. This result coincides with the findings of Delmas and Gergaud (Reference Delmas and Gergaud2014) on taking environmental measures. The control variables were positively influenced by sales and number of workers per square metre. However, size, measured by the number of workers and area cultivated by the family farm, has a negative influence on water use efficiency, though less intensely and significantly than age.
The inherit variable was entered in Model 2 and shows that the intention of leaving the family farm to the next generation in inheritance has a strong positive impact on the effort to use water more efficiently. This result is in agreement with those of other studies such as the one by Delmas and Gergaud (Reference Delmas and Gergaud2014). Compared to Model 1 as the control, there was a 3.26% increase in explained variance in Model 2, and the model as a whole is significant.
The variables related to decision-makers were then added in Model 3. This reveals that family farms attempt to be more efficient in the use of water when there are fewer people making decisions, when the decision-makers are under 45 (as in Arcury and Christianson, Reference Arcury and Christianson1990; Tang et al., Reference Tang, Folmer and Xue2013), and above all, when they are women. Contrary to expectations, education is hardly significant, and its influence is negative, as in Spriggs et al. (Reference Spriggs, Yu, Deeds and Sorenson2012). This model had a moderate increase of 3.9% in explained variance and is still significant as a whole.
Model 4 incorporates variables related to awareness of the environment and of water usage in the family farm, and some changes appear in the estimated coefficients of the variables entered in the three previous steps. In this case, the generation variable had a negative influence (as in Spriggs et al., Reference Spriggs, Yu, Deeds and Sorenson2012) as did the number of workers per square metre. This means that the family farm strives to use water more efficiently when it is younger, smaller, less intensive in work and sales per square metre, earns more income, and is going to be inherited. The educational level of the decision-makers has a positive influence and gains in significance. Therefore, according to Model 4, family farms use water more efficiently when fewer people are making decisions, they are younger, more educated and female. The latter variable has a higher weight and is more significant. With regard to new variables incorporated, family farms take more measures for efficient water use when they display greater awareness of water and environmental issues (as in Tang et al., Reference Tang, Folmer and Xue2013), specifically when (i) they have a larger amount of IPM per square metre; (ii) they receive government subsidies for it; (iii) their reasons are economic (cost saving and/or increased sales) as well as environmental; (iv) they believe that the other family farms in the sector are doing the same, (v) and the farmer also tries to reduce water use in his daily life. Amongst all variables, the most determining and significant one is having received government aid, followed closely by the number of women who make decisions in the family farm and thirdly the fact that the farm is to be inherited. Model 4 implies an important increase in the variance explained (7.6%) and a fit of up to almost 84%.
DISCUSSION
The water shortage in Spain is partly due to poor management of the resource in the agricultural sector. It is therefore important to ascertain which factors have an impact on water efficiency in this productive activity. In the present study, we have analysed how certain characteristics of a family farm and its decision-makers influence the extent of its awareness of efficient water use, taking as a single case reference family farms in the fruit and vegetables sector in Almería, Spain. The data from the surveys carried out have been analysed and a hierarchical regression model was estimated.
Characteristics of family farms
According to our results, the family farms that are most aware and strive to use water more efficiently have the following characteristics: (i) They are younger and have been in the hands of fewer generations. Results on the influence on the family farm's behaviour of this generational participation are similar to those of Ling and Kellermanns (Reference Ling and Kellermanns2010) on the ease in making decisions, and Spriggs et al. (Reference Spriggs, Yu, Deeds and Sorenson2012) on capacity for innovation, although this is not the case in Delmas and Gergaud (Reference Delmas and Gergaud2014) on environmental measures in general; (ii) They are smaller, with fewer workers and a smaller area cultivated. These results agree with the findings of McGrath (Reference McGrath2001) on capacity for innovation in business; Grant et al. (Reference Grant, Jones and Bergesen2002) on their polluting activities; and Berrone et al. (Reference Berrone, Cruz, Gómez Mejía and Larraza Kintana2010) and Delmas and Gergaud (Reference Delmas and Gergaud2014) on environmental measures. The ethical, social and environmental values of the farmer/decision-maker of small and medium-sized farms are usually important in defining their mission and strategy (Delmas and Gergaud, Reference Delmas and Gergaud2014); (iii) They intend to leave the farm to the next generation. This is the case of 90% of the farms surveyed, since all of them are family businesses. This factor has a very strong impact and shows that the owners who intend to leave their farms to their children, taken on a longer term view, are more receptive to the needs of future generations and the sustainability of their family farms. The literature on stakeholders has demonstrated how businesses have to respond to the pressures of these groups by adopting ecological practices (Delmas and Toffel, Reference Delmas and Toffel2004). However, this framework usually ignores family farms and the connections they have with the future of their own family members (with exceptions, such as Bingham et al. Reference Bingham, Dyer, Smith and Adams2011). Our contribution to this literature is to show that future generations should be considered as the main stakeholders, since their existence influences the decisions of the farmers on general environmental matters and water in particular (corroborating the results found by Delmas and Gergaud, Reference Delmas and Gergaud2014); (iv) There are fewer decision-makers, as is usually the case in smaller farms (see also Berrone et al., Reference Berrone, Cruz, Gómez Mejía and Larraza Kintana2010; Spriggs et al., Reference Spriggs, Yu, Deeds and Sorenson2012); they are younger (as in Arcury and Christianson, Reference Arcury and Christianson1990; Berrone et al., Reference Berrone, Cruz, Gómez Mejía and Larraza Kintana2010; Tang et al., Reference Tang, Folmer and Xue2013); and they are better educated (as in Jones and Dunlap, Reference Jones and Dunlap1992); (v) They have more women amongst their decision-makers. This is one of the most significant and influential variables. Only 36% of all decision-makers in these family farms are women and only 40% of those who form part of the family farm participate in their decisions. Some studies have analysed the contribution of women to different types of strategic decision-making in business (Farmar-Bowers, Reference Farmar-Bowers2010; Grubbström et al., Reference Grubbström, Stenbacka and Joosse2014). However, the difference in environmental awareness between men and women is not usually studied; (vi) They have more IPM and other certifications of environmental quality of production. The concept of IPM is founded on a more sustainable approach based on the criteria of good agricultural practices, implying the efficient use of means and factors. In fact, it fosters, for example, irrigation techniques that promote water savings; (vii) They receive government subsidies to implant these measures. This is the most determining and significant variable. Family farms that have received government aid are more aware and have taken the most water efficiency measures (see Karali et al., Reference Karali, Brunner, Doherty, Hersperger and Rounsevell2014, for general environmental measures). In addition, we can observe economic motivations, such as reducing costs or increasing sales, should also be taken into account. Most family farms take water efficiency measures to decrease the costs associated with water use. We can therefore conclude that in addition to being desirable for sustainability, this type of measure should also be profitable for the family farm (Kienzler et al., Reference Kienzler, Lamers, McDonald, Mirzabaeva, Ibragimov, Egamberdiev, Ruzibaev and Akramkhanov2012). They consider the other companies in the sector to be aware of efficient water use. Competitors are one of the most common interest groups, or stakeholders, of businesses (Freeman, Reference Freeman1984). Our study corroborates that the awareness of other family farms in the sector of efficient water use positively influences the family farm's water strategy (as in Liu et al., Reference Liu, Yu, Bi and Zhang2009, for general environmental measures).
How to improve water use efficiency?
According to the results expounded, some influential drivers can be identified as key factors when designing agri-environmental-water measures. It is important for governments to promote the creation of new farms and renew instruments and technologies related to water use. Moreover, policy makers should promote family farms, as the presence of a successor generally makes the farm more receptive to the needs of future generations and sustainability. This driver can be seen as an opportunity for policy-makers to encourage a greater involvement of the successor in the decision-making process. Although women contribute significantly to society's knowledge and sensitivity from different spheres, their influence is not sufficiently visible because they are not adequately represented in discussion forums or decision-making circles. Policies that promote the access of women to management, both by teaching equality in schools and by measures that allow family and work commitments to be reconciled, could therefore have the externality benefit of bringing more farms into water efficiency.
There is no doubt that family farms in Almería have received scant government assistance, as only 11% of the sample received financial aid, most of which derived from European funding. This kind of aid is very scant in Spain, especially since the economic crisis, and most of it comes from the Common Agricultural Policy (CAP) via Operative Programs of the Fruit and Vegetable sector. In these subsidy programmes, investment should be made 1 year previously and only half of the amount is subsidized (with a limit of 4% of the farmer's sales), so not all farmers can invest. In addition, this aid goes to all types of infrastructure and it does not only focus on the use of water. If there is no major public investment, water shortages cannot be solved by the farmers’ efficient water usage alone. In this sense, the low public budgets in Spain due to the recession are an additional problem. It would be recommendable for governments at all levels to provide more support to these family farms through agro-environmental lines, defraying part of the expenses derived from certification and implanting water efficiency measures, for instance, by providing personalized advisory services to growers.
On the other hand, cooperatives play a key role in the development of respectful environmental actions. Most farms are family-farms and therefore cooperatives, as producer organizations in an intermediate position within the food system, should aim to introduce environmental practices including efficiency in water use, covering the whole process in the home market. They offer advice on investment aid and on new irrigation technologies and methods. Therefore, they are certainly a factor that should be taken into account in studies of this type and they were originally intended for inclusion. However, 85% of the surveyed farmers (similar to the average of farmers in the sector) are part of a production, purchase or consumption cooperative, so it is hardly a differentiable variable. Nonetheless, it is true that future policies should focus on the cooperative aspect for saving water.
Neighbouring farmers’ experiences about water efficiency measures can influence other farmers’ behaviour. Workshops and other activities showing the advantages and results obtained with these measures could be developed to promote the benefits of water efficient use. Policy makers could even take advantage of their efficient water usage to act as advocates for farms located in other areas. Finally, we insist on the need for government institutions to develop environmental awareness policies in schools and directed at both farmers and the population as a whole.
In general, the study provides evidence regarding the influences of family farming features and behaviour on water efficiency that may prove useful for analyses on agriculture, and particularly those concerning family farming systems. This evidence could be used in other Mediterranean regions with an important fruit and vegetables sector and similar characteristics, such as the south of Europe (e.g., south of Italy, Greece or Turkey, or Israel) and/or the north of Africa. Or even in other non-Mediterranean areas such as the south of Morocco (e.g., Agadir). This research is not exempt from limitations, and overcoming them would provide research lines for future studies. First, the analysis was limited to the agricultural sector in the province of Almería. As explained, this is a particular case in which the productive structure is based on small family farms. It would therefore be of interest to explore similar matters in other more international contexts or even in other farming sectors. Second, the data was concentrated on water use efficiency, but future work could also include other sustainable practices. Additionally, the surveys collected data on variables at a specific moment in time. A longitudinal analysis would determine whether the relationships identified in this study persist over time.
Acknowledgements
This research was partially funded by Spanish MCINN and FEDER aid (project ECO2014-52268) and by Andalusian Regional Government (project SEJ-5827, Consejería de Economía, Innovación y Ciencia).
